Data Science for Economists

Software requirements

External software

Today we’ll be using SelectorGadget, which is a Chrome extension that makes it easy to discover CSS selectors. (Install the extension directly here.) Please note that SelectorGadget is only available for Chrome. If you prefer using Firefox, then you can try ScrapeMate.

R packages

New: rvest, janitor
Already used: tidyverse, lubridate, hrbrthemes

Recall that rvest was automatically installed with the rest of the tidyverse. However, these lecture notes assume that you have rvest 1.0.0, which — at the time of writing — has to installed as the development version from GitHub. The code chunk below should take care of installing (if necessary) and loading the packages that you need for today’s lecture.

## Install development version of rvest if necessary
if (numeric_version(packageVersion("rvest")) < numeric_version('0.99.0')) {
  remotes::install_github('tidyverse/rvest')
}
## Load and install the packages that we'll be using today
if (!require("pacman")) install.packages("pacman")
pacman::p_load(tidyverse, rvest, lubridate, janitor, hrbrthemes)
## My preferred ggplot2 plotting theme (optional)
theme_set(hrbrthemes::theme_ipsum())

Webscraping basics

The next two lectures are about getting data, or “content”, off the web and onto our computers. We’re all used to seeing this content in our browers (Chrome, Firefox, etc.). So we know that it must exist somewhere. However, it’s important to realise that there are actually two ways that web content gets rendered in a browser:

Server-side
Client side

You can read here for more details (including example scripts), but for our purposes the essential features are as follows:

1. Server-side

The scripts that “build” the website are not run on our computer, but rather on a host server that sends down all of the HTML code.
- E.g. Wikipedia tables are already populated with all of the information — numbers, dates, etc. — that we see in our browser.
In other words, the information that we see in our browser has already been processed by the host server.
You can think of this information being embeded directly in the webpage’s HTML.
Webscraping challenges: Finding the correct CSS (or Xpath) “selectors”. Iterating through dynamic webpages (e.g. “Next page” and “Show More” tabs).
Key concepts: CSS, Xpath, HTML

2. Client-side

The website contains an empty template of HTML and CSS.
- E.g. It might contain a “skeleton” table without any values.
However, when we actually visit the page URL, our browser sends a request to the host server.
If everything is okay (e.g. our request is valid), then the server sends a response script, which our browser executes and uses to populate the HTML template with the specific information that we want.
Webscraping challenges: Finding the “API endpoints” can be tricky, since these are sometimes hidden from view.
Key concepts: APIs, API endpoints

Over the next week, we’ll use these lecture notes — plus some student presentations — to go over the main differences between the two approaches and cover the implications for any webscraping activity. I want to forewarn you that webscraping typically involves a fair bit of detective work. You will often have to adjust your steps according to the type of data you want, and the steps that worked on one website may not work on another. (Or even work on the same website a few months later). All this is to say that webscraping involves as much art as it does science.

The good news is that both server-side and client-side websites allow for webscraping.¹ If you can see it in your browser, you can scrape it.

Caveat: Ethical and legal limitations

The previous sentence elides some important ethical and legal considerations. Just because you can scrape it, doesn’t mean you should. It is ultimately your responsibility to determine whether a website maintains legal restrictions on the content that it provides. Similarly, the tools that we’ll be using are very powerful. It’s fairly easy to write up a function or program that can overwhelm a host server or application through the sheer weight of requests. A computer can process commands much, much faster than we can ever type them up manually. We’ll return to the “be nice” mantra at the end of this lecture, as well as in the next lecture.

Webscraping with rvest (server-side)

The primary R package that we’ll be using today is rvest (link), a simple webscraping library inspired by Python’s Beautiful Soup (link), but with extra tidyverse functionality. rvest is designed to work with webpages that are built server-side and thus requires knowledge of the relevant CSS selectors… Which means that now is probably a good time for us to cover what these are.

Student presentation: CSS and SelectorGadget

Time for a student presentation on CSS (i.e Cascading Style Sheets) and SelectorGadget. Click on the links if you are reading this after the fact. In short, CSS is a language for specifying the appearance of HTML documents (including web pages). It does this by providing web browsers a set of display rules, which are formed by:

Properties. CSS properties are the “how” of the display rules. These are things like which font family, styles and colours to use, page width, etc.
Selectors. CSS selectors are the “what” of the display rules. They identify which rules should be applied to which elements. E.g. Text elements that are selected as “.h1” (i.e. top line headers) are usually larger and displayed more prominently than text elements selected as “.h2” (i.e. sub-headers).

The key point is that if you can identify the CSS selector(s) of the content you want, then you can isolate it from the rest of the webpage content that you don’t want. This where SelectorGadget comes in. We’ll work through an extended example (with a twist!) below, but I highly recommend looking over this quick vignette before proceding.

Application: Mens 100 meters (Wikipedia)

Okay, let’s get to an application. Say that we want to scrape the Wikipedia page on the Men’s 100 metres world record progression.

First, open up this page in your browser. Take a look at its structure: What type of objects does it contain? How many tables does it have? Do these tables all share the same columns? What row- and columns-spans? Etc.

Once you’ve familiarised yourself with the structure, read the whole page into R using the rvest::read_html() function.

# library(rvest) ## Already loaded

m100 = read_html("http://en.wikipedia.org/wiki/Men%27s_100_metres_world_record_progression") 
m100

## {html_document}
## <html class="client-nojs" lang="en" dir="ltr">
## [1] <head>\n<meta http-equiv="Content-Type" content="text/html; charset=UTF-8 ...
## [2] <body class="mediawiki ltr sitedir-ltr mw-hide-empty-elt ns-0 ns-subject  ...

As you can see, this is an XML document² that contains everything needed to render the Wikipedia page. It’s kind of like viewing someone’s entire LaTeX document (preamble, syntax, etc.) when all we want are the data from some tables in their paper.

Table 1: Pre-IAAF (1881–1912)

Let’s try to isolate the first table on the page, which documents the unofficial progression before the IAAF. The core function that we’ll use to isolate and extract this table from the rest of the HTML document is rvest::html_element(). I’ll use SelectorGadget to identify the table’s CSS selector (in this case I get “div+ .wikitable :nth-child(1)”). Finally, we should then be able to convert the scraped table into a data frame using rvest::html_table(). I’ll assign this object as pre_iaaf.

pre_iaaf = 
  m100 %>%
  html_element("div+ .wikitable :nth-child(1)") %>% ## select table element
  html_table()                                      ## convert to data frame

pre_iaaf

## # A tibble: 21 x 5
##     Time Athlete            Nationality    `Location of races`   Date           
##    <dbl> <chr>              <chr>          <chr>                 <chr>          
##  1  10.8 Luther Cary        United States  Paris, France         July 4, 1891   
##  2  10.8 Cecil Lee          United Kingdom Brussels, Belgium     September 25, …
##  3  10.8 Étienne De Ré      Belgium        Brussels, Belgium     August 4, 1893 
##  4  10.8 L. Atcherley       United Kingdom Frankfurt/Main, Germ… April 13, 1895 
##  5  10.8 Harry Beaton       United Kingdom Rotterdam, Netherlan… August 28, 1895
##  6  10.8 Harald Anderson-A… Sweden         Helsingborg, Sweden   August 9, 1896 
##  7  10.8 Isaac Westergren   Sweden         Gävle, Sweden         September 11, …
##  8  10.8 Isaac Westergren   Sweden         Gävle, Sweden         September 10, …
##  9  10.8 Frank Jarvis       United States  Paris, France         July 14, 1900  
## 10  10.8 Walter Tewksbury   United States  Paris, France         July 14, 1900  
## # … with 11 more rows

Great, it worked!

I’ll tidy things up a bit so that the data frame is easier to work with in R. First, I’ll use the janitor::clean_names() convenience function to remove spaces and capital letters from the column names. (Q: How else could we have done this?) Second, I’ll use the lubridate::mdy() function to convert the date string to a format that R actually understands.

# library(janitor) ## Already loaded
# library(lubridate) ## Already loaded
 
pre_iaaf =
  pre_iaaf %>%
  clean_names() %>%         ## fix the column names
  mutate(date = mdy(date))  ## convert string to date format

pre_iaaf

## # A tibble: 21 x 5
##     time athlete               nationality    location_of_races       date      
##    <dbl> <chr>                 <chr>          <chr>                   <date>    
##  1  10.8 Luther Cary           United States  Paris, France           1891-07-04
##  2  10.8 Cecil Lee             United Kingdom Brussels, Belgium       1892-09-25
##  3  10.8 Étienne De Ré         Belgium        Brussels, Belgium       1893-08-04
##  4  10.8 L. Atcherley          United Kingdom Frankfurt/Main, Germany 1895-04-13
##  5  10.8 Harry Beaton          United Kingdom Rotterdam, Netherlands  1895-08-28
##  6  10.8 Harald Anderson-Arbin Sweden         Helsingborg, Sweden     1896-08-09
##  7  10.8 Isaac Westergren      Sweden         Gävle, Sweden           1898-09-11
##  8  10.8 Isaac Westergren      Sweden         Gävle, Sweden           1899-09-10
##  9  10.8 Frank Jarvis          United States  Paris, France           1900-07-14
## 10  10.8 Walter Tewksbury      United States  Paris, France           1900-07-14
## # … with 11 more rows

Now that we have our cleaned pre-IAAF data frame, we could easily plot it. I’m going to hold off doing that until we’ve scraped the rest of the WR data. But first, an aside on browser inspection tools.

Aside: Get CSS selectors via browser inspection tools

SelectorGadget is a great tool. But it isn’t available on all browsers and can involve more work than I’d like sometimes with all of the additional clicking.³ I therefore wanted to mention an alternative (and very precise) approach to obtaining CSS selectors: Use the “inspect web element” feature of your browser.

Here’s a quick example using Google Chrome. First, I open up the inspect console (Ctrl+Shift+I, or right-click and choose “Inspect”). I then proceed to scroll over the source elements, until Chrome highlights the table of interest on the actual page. Once the table (or other element of interest) is highlighted, I can grab its CSS by right-clicking and selecting Copy -> Copy selector.

In general, I prefer to obtain CSS selectors using this “inspect” method with my browser. But each to their own.

Challenge

Your turn: Download the next two tables from the same WR100m page. Combine these two new tables with the one above into a single data frame and then plot the record progression. Answer below. (No peeking until you have tried first.)

Table 2: Pre-automatic timing (1912–1976)

Let’s start with the second table.

iaaf_76 =
  m100 %>%
  html_element("#mw-content-text > div > table:nth-child(14)") %>%
  html_table()

As we did with the previous table, let’s fix the column names and coerce the date string to a format that R understands.

iaaf_76 =
  iaaf_76 %>%
  clean_names() %>%
  mutate(date = mdy(date)) 

iaaf_76

## # A tibble: 54 x 8
##     time wind   auto athlete     nationality location_of_race   date       ref  
##    <dbl> <chr> <dbl> <chr>       <chr>       <chr>              <date>     <chr>
##  1  10.6 ""     NA   Donald Lip… United Sta… Stockholm, Sweden  1912-07-06 [2]  
##  2  10.6 ""     NA   Jackson Sc… United Sta… Stockholm, Sweden  1920-09-16 [2]  
##  3  10.4 ""     NA   Charley Pa… United Sta… Redlands, USA      1921-04-23 [2]  
##  4  10.4 "0.0"  NA   Eddie Tolan United Sta… Stockholm, Sweden  1929-08-08 [2]  
##  5  10.4 ""     NA   Eddie Tolan United Sta… Copenhagen, Denma… 1929-08-25 [2]  
##  6  10.3 ""     NA   Percy Will… Canada      Toronto, Canada    1930-08-09 [2]  
##  7  10.3 "0.4"  10.4 Eddie Tolan United Sta… Los Angeles, USA   1932-08-01 [2]  
##  8  10.3 ""     NA   Ralph Metc… United Sta… Budapest, Hungary  1933-08-12 [2]  
##  9  10.3 ""     NA   Eulace Pea… United Sta… Oslo, Norway       1934-08-06 [2]  
## 10  10.3 ""     NA   Chris Berg… Netherlands Amsterdam, Nether… 1934-08-26 [2]  
## # … with 44 more rows

Table 3: Modern Era (1977 onwards)

For the final table, I’ll just run the code all at once. By now you should recognise all of the commands.

iaaf =
  m100 %>%
  html_element("#mw-content-text > div > table:nth-child(19)") %>%
  html_table() %>%
  clean_names() %>%
  mutate(date = mdy(date))

iaaf

## # A tibble: 24 x 8
##     time wind   auto athlete nationality location_of_race date      
##    <dbl> <chr> <dbl> <chr>   <chr>       <chr>            <date>    
##  1 10.1  1.3      NA Bob Ha… United Sta… Tokyo, Japan     1964-10-15
##  2 10.0  0.8      NA Jim Hi… United Sta… Sacramento, USA  1968-06-20
##  3 10.0  2.0      NA Charle… United Sta… Mexico City, Me… 1968-10-13
##  4  9.95 0.3      NA Jim Hi… United Sta… Mexico City, Me… 1968-10-14
##  5  9.93 1.4      NA Calvin… United Sta… Colorado Spring… 1983-07-03
##  6  9.83 1.0      NA Ben Jo… Canada      Rome, Italy      1987-08-30
##  7  9.93 1.0      NA Carl L… United Sta… Rome, Italy      1987-08-30
##  8  9.93 1.1      NA Carl L… United Sta… Zürich, Switzer… 1988-08-17
##  9  9.79 1.1      NA Ben Jo… Canada      Seoul, South Ko… 1988-09-24
## 10  9.92 1.1      NA Carl L… United Sta… Seoul, South Ko… 1988-09-24
## # … with 14 more rows, and 1 more variable: notes_note_2 <chr>

Combined eras

Let’s bind all these separate eras into a single data frame. I’ll use dplyr:: bind_rows() again and select in the common variables only. I’ll also add a new column describing which era an observation falls under.

wr100 = 
  rbind(
    pre_iaaf %>% select(time, athlete, nationality, date) %>% mutate(era = "Pre-IAAF"),
    iaaf_76 %>% select(time, athlete, nationality, date) %>% mutate(era = "Pre-automatic"),
    iaaf %>% select(time, athlete, nationality, date) %>% mutate(era = "Modern")
    )

wr100

## # A tibble: 99 x 5
##     time athlete               nationality    date       era     
##    <dbl> <chr>                 <chr>          <date>     <chr>   
##  1  10.8 Luther Cary           United States  1891-07-04 Pre-IAAF
##  2  10.8 Cecil Lee             United Kingdom 1892-09-25 Pre-IAAF
##  3  10.8 Étienne De Ré         Belgium        1893-08-04 Pre-IAAF
##  4  10.8 L. Atcherley          United Kingdom 1895-04-13 Pre-IAAF
##  5  10.8 Harry Beaton          United Kingdom 1895-08-28 Pre-IAAF
##  6  10.8 Harald Anderson-Arbin Sweden         1896-08-09 Pre-IAAF
##  7  10.8 Isaac Westergren      Sweden         1898-09-11 Pre-IAAF
##  8  10.8 Isaac Westergren      Sweden         1899-09-10 Pre-IAAF
##  9  10.8 Frank Jarvis          United States  1900-07-14 Pre-IAAF
## 10  10.8 Walter Tewksbury      United States  1900-07-14 Pre-IAAF
## # … with 89 more rows

All that hard works deserves a nice plot, don’t you think?

wr100 %>%
  ggplot(aes(x=date, y=time, col=fct_reorder2(era, date, time))) + 
  geom_point(alpha = 0.7) +
  labs(
    title = "Men's 100m world record progression",
    x = "Date", y = "Time",
    caption = "Source: Wikipedia"
    ) +
  theme(legend.title = element_blank()) ## Switch off legend title

Summary

Web content can be rendered either 1) server-side or 2) client-side.
To scrape web content that is rendered server-side, we need to know the relevant CSS selectors.
We can find these CSS selectors using SelectorGadget or, more precisely, by inspecting the element in our browser.
We use the rvest package to read into the HTML document into R and then parse the relevant nodes.
- A typical workflow is: read_html(URL) %>% html_element(CSS_SELECTORS) %>% html_table().
- You might need other functions depending on the content type (e.g. see ?html_text).
Just because you can scrape something doesn’t mean you should (i.e. ethical and legal restrictions).
Webscraping involves as much art as it does science. Be prepared to do a lot of experimenting and data cleaning.
Next lecture: Webscraping: (2) Client-side and APIs.

Further resources and exercises

In the next lecture, we’re going to focus on client-side web content and interacting with APIs. For the moment, you can practice your rvest-based scraping skills by following along with any of the many (many) tutorials available online. I want to make two particular suggestions, though:

More on web ettiquette

We spoke a bit about the “be nice” scraping motto at the beginning of the lecture. I also wanted to point you to the polite package (link). It provides some helpful tools to maintain web etiquette, such as checking for permission and not hammering the host website with requests. It also plays very nicely with the rvest workflow that we covered today, so please take a look.

Modeling and prediction

We’ll get to analysis section of the course (regression, etc.) next week. However, today’s dataset provides a good platform to start thinking about these issues. How would you model the progression of the Men’s 100 meter record over time? For example, imagine that you had to predict today’s WR in 2005. How do your predictions stack up against the actual record (i.e. Usain Bolt’s 9.58 time set in 2009)? How do you handle rescinded times? How do you intepret all of this?

Hint: See the ?broom::tidy() help function for extracting regression coefients in a convenient data frame. We’ve already seen the geom_smooth() function, but for some nice tips on (visualizing) model predictions, see Chap. 23 of the R4DS book, or Chap. 6.4 of the SocViz book. The generic base::predict() function has you covered, although the tidyverse’s modelr package has some nice wrapper functions that you will probably find useful for this suggested exercise.

As we’ll see during the next lecture, scraping a website or application that is built on a client-side (i.e. API) framework is often easier; particularly when it comes to downloading information en masse.↩︎
XML stands for Extensible Markup Language and is one of the primary languages used for encoding and formatting web pages.↩︎
Historically, at least, it also had a tendency to provide CSS selectors that weren’t exactly what we were looking for. To be fair, this may have reflected some issues coming from the R + rvest as much as anything else. These minor incompatibilities have been largely eliminated with rvest 1.0.0… prompting a re-write of these notes!↩︎